RVPO: Risk-Sensitive Alignment via Variance Regularization
Current critic-less RLHF methods aggregate multi-objective rewards via an arithmetic mean, leaving them vulnerable to constraint neglect: high-magnitude success in one objective can numerically offset critical failures in others (e.g., safety or formatting), masking low-performing “bottleneck” rewar...
Velox: Learning Representations of 4D Geometry and Appearance
We introduce a framework for learning latent representations of 4D objects which are descriptive, faithfully capturing object geometry and appearance; compressive, aiding in downstream efficiency; and accessible, requiring minimal input, i.e., an unstructured dynamic point cloud, to construct. Speci...
Endogenous Regime Switching Driven by Scalar-Irreducible Learning Dynamics
arXiv:2605.04054v1 Announce Type: new
Abstract: Achieving endogenous regime switching is crucial for the emergence of autonomous intelligence, yet remains a central challenge for existing machine learning frameworks, where such transitions are typically externally imposed. In this work, we introduc...
Transformation Categorization Based on Group Decomposition Theory Using Parameter Division
arXiv:2605.04056v1 Announce Type: new
Abstract: Representation learning seeks meaningful sensory representations without supervision and can model aspects of human development. Although many neural networks empirically learn useful features, a principled account of what makes a representation "good...
Structured Progressive Knowledge Activation for LLM-Driven Neural Architecture Search
arXiv:2605.04057v1 Announce Type: new
Abstract: This paper focuses on a key challenge in Neural Architecture Search (NAS): integrating established architectural knowledge while exploring new designs under expensive evaluations. Large language models (LLMs) are a promising assistant for NAS because ...
MP-ISMoE: Mixed-Precision Interactive Side Mixture-of-Experts for Efficient Transfer Learning
arXiv:2605.04058v1 Announce Type: new
Abstract: Parameter-efficient transfer learning (PETL) has emerged as a pivotal paradigm for adapting pre-trained foundation models to downstream tasks, significantly reducing trainable parameters yet suffering from substantial memory overhead caused by gradien...
arXiv:2605.04050v1 Announce Type: new
Abstract: We introduce Lossless Context Management (LCM), a deterministic architecture for LLM memory that outperforms Claude Code on long-context tasks. When benchmarked using Opus 4.6, our LCM-augmented coding agent, Volt, achieves higher scores than Claude C...
Actionable Real-Time Modeling of Surgical Team Dynamics via Time-Expanded Interaction Graphs
arXiv:2605.04169v1 Announce Type: new
Abstract: Surgical team performance arises from complex interactions between technical execution and non-technical skills, including communication and coordination dynamics. However, current surgical AI systems predominantly model visual workflow signals, lacki...
Pro$^2$Assist: Continuous Step-Aware Proactive Assistance with Multimodal Egocentric Perception for Long-Horizon Procedural Tasks
arXiv:2605.04227v1 Announce Type: new
Abstract: Procedural tasks with multiple ordered steps are ubiquitous in daily life. Recent advances in multimodal large language models (MLLMs) have enabled personal assistants that support daily activities. However, existing systems primarily provide reactive...
Jensen Huang called it "the ChatGPT moment for robotics." Deloitte says 80% of businesses plan to use physical AI within two years. Here is what you actually need to know, and do, to prepare…
StateSMix: Online Lossless Compression via Mamba State Space Models and Sparse N-gram Context Mixing
arXiv:2605.02904v1 Announce Type: new
Abstract: We present StateSMix, a fully self-contained lossless compressor that couples an online-trained Mamba-style State Space Model (SSM) with sparse n-gram context mixing and arithmetic coding. The model is initialised from scratch and trained token-by-tok...
eOptShrinkQ: Near-Lossless KV Cache Compression Through Optimal Spectral Denoising and Quantization
arXiv:2605.02905v1 Announce Type: new
Abstract: We show that the key-value (KV) cache in transformer attention heads admits a natural decomposition into a low-rank \emph{shared context} component and a full-rank \emph{per-token} residual, well described by the spiked random matrix model. This obser...
An End-to-End Framework for Building Large Language Models for Software Operations
arXiv:2605.02906v1 Announce Type: new
Abstract: In the field of software operations, Large Language Models (LLMs) have attracted increasing attention. However, existing research has not yet achieved efficient and effective end-to-end intelligent operations due to low-quality data, fragmented knowle...
Delay, Plateau, or Collapse: Evaluating the Impact of Systematic Verification Error on RLVR
arXiv:2605.02909v1 Announce Type: new
Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) has become a powerful approach for improving the reasoning capabilities of large language models (LLMs). While RLVR is designed for tasks with verifiable ground-truth answers, real-world verifiers ...
CreativityBench: Evaluating Agent Creative Reasoning via Affordance-Based Tool Repurposing
arXiv:2605.02910v2 Announce Type: new
Abstract: Recent advances in large language models have led to strong performance on reasoning and environment-interaction tasks, yet their ability for creative problem-solving remains underexplored. We study this capability through the lens of creative tool us...
Stable Agentic Control: Tool-Mediated LLM Architecture for Autonomous Cyber Defense
arXiv:2605.03034v1 Announce Type: new
Abstract: Agentic systems involved in high-stake decision-making under adversarial pressure need formal guarantees not offered by existing approaches. Motivated by the operational needs of security operations centers (SOCs) that must configure endpoint detectio...
Programmatic Context Augmentation for LLM-based Symbolic Regression
arXiv:2605.03101v1 Announce Type: new
Abstract: Symbolic regression (SR), the task of discovering mathematical expressions that best describe a given dataset, remains a fundamental challenge in scientific discovery. Traditional approaches, primarily based on genetic algorithms and related evolution...
AI lets chemists design molecules by simply describing them
Creating complex molecules usually requires years of experience and countless decisions, but a new AI system is changing that. Synthegy lets chemists guide synthesis and reaction planning using simple language, while powerful algorithms generate and evaluate possible solutions. The AI doesn’t just c...
From Where Things Are to What They’re For: Benchmarking Spatial–Functional Intelligence for Multimodal LLMs
True spatial intelligence for multimodal agents transcends low-level geometric perception, evolving from knowing where things are to understanding what they are for. While existing benchmarks, such as VSI-Bench, effectively evaluate this foundational geometric stage, they fall short of probing the h...
Normalizing Flows (NFs) are a classical family of likelihood-based methods that have received revived attention. Recent efforts such as TARFlow have shown that NFs are capable of achieving promising performance on image modeling tasks, making them viable alternatives to other methods such as diffusi...
Microsoft at NSDI 2026: Advances in large-scale networked systems
Microsoft researchers share advances in building and operating large-scale distributed systems, spanning datacenters, networking, and the growing intersection with AI during NSDI ’26.
The post Microsoft at NSDI 2026: Advances in large-scale networked systems appeared first on Microsoft Research.
Agentopic: A Generative AI Agent Workflow for Explainable Topic Modeling
arXiv:2605.00833v1 Announce Type: new
Abstract: Agentopic is a novel agent-based workflow for explainable topic modeling that leverages the reasoning capabilities of Large Language Models (LLMs). Existing topic modeling approaches such as Latent Dirichlet Allocation (LDA) and BERTopic often lack tr...